The Effect of Empiric Initiation of Treatment on Outcomes Among Patients Presenting With Extrapulmonary Mycobacterium Tuberculosis Infection in a Community Hospital Setting.

Background: Extrapulmonary tuberculosis (EPTB) accounts for ~20% of Mycobacterium tuberculosis infections in Pakistan. Diagnostic delays are common and may worsen outcomes. We evaluated whether empiric treatment initiation improves outcomes among patients with EPTB at a large public hospital in Islamabad, Pakistan. Methods: Retrospective cohort of consecutive EPTB cases at Pakistan Institute of Medical Sciences (January-June 2016). Patients were categorized as clinically diagnosed (symptoms/radiology) or laboratory-supported (histology, microbiology, or biochemical tests). Standard Directly Observed Therapy Short-course (DOTS) based regimens were used. Outcomes were treatment success, failure, loss to follow-up, and mortality; groups were compared using chi-square testing. Results: Of 317 EPTB patients, 275 completed treatment and were analyzed. Mean age was 34.4 years; 57% were male. The most common sites were pleural (28.7%) and lymphatic (20.3%) TB. Clinically diagnosed patients had higher cure rates than laboratory supported diagnosis group patients (87.6% vs. 79.6%; p = 0.221) and significantly lower mortality (5.3% vs. 9.3%; p = 0.041). Among patients who completed treatment, 82.9% achieved treatment success; overall cure rate was 71% including non-completion and loss to follow-up. Conclusions: Early empiric therapy may be associated with improved survival and higher cure rates compared with delaying treatment pending laboratory confirmation. Context-appropriate EPTB guidelines emphasizing prompt clinical treatment are warranted in high burden settings.